James Hammitt is Professor of Economics and Decision Sciences at Harvard School of Public Health. He is the director of the Harvard Center for Risk Analysis and fellow of the Society for Risk Analysis. His research concerns the development and application of quantitative methods—including benefit-cost, decision and risk analysis—to health and environmental policy. He holds a PhD in Public Policy from Harvard University. Benefit-cost analysis (BCA) is a form of applied welfare analysis that is often used to help choose government policies, especially for environmental, health, and safety regulation and investment in transportation. It is often required in the United States, and is becoming more common in France and the European Union.
What is the goal of BCA? What question does it answer? These questions are critical for understanding how BCA results should be interpreted and how BCA should be conducted. There are at least two possible bases for justifying BCA, positive and normative. The positive basis derives from the text-book description of BCA as a method that identifies policy changes that satisfy the Kaldor-Hicks compensation test. The test asks whether those who benefit from a policy change could compensate those who are harmed so that everyone would judge himself better off with the policy change and compensation payments than without. This justification leaves unanswered the normative question of whether passing the Kaldor-Hicks compensation test is either a necessary or sufficient condition for the policy change to improve social wellbeing. The normative basis asserts that BCA is a method to identify policy changes that constitute social improvements, where improvement must be defined in some way that is external to BCA. Several normative justifications can be offered. One is derived from a form of utilitarianism in which the objective is to maximize the sum of well-being in the society and it is assumed that well-being can be measured using standard economic concepts such as compensating and equivalent variation. This justification leads naturally to the suggestion that the monetary values of benefits and costs should be weighted depending on whether they fall on rich or poor, given the intuition that marginal utility declines with wealth. Other normative justifications are motivated by pragmatism. One such claim is that BCA is a practical method that approximates the result of an ideal but impractical measure of social welfare. Another is that BCA helps promote consistent decision making by avoiding random errors and protecting against cognitive mistakes that can arise when a decision maker tries to evaluate a policy change by holistic judgment. A third claim is that, if policies that maximize net benefits are routinely chosen, everyone will be better off in the long run because those who benefit and those who are harmed will tend to vary across decisions. Obviously, the strength of these pragmatic claims depends on the alternatives with which BCA is compared. Conflict between positive and normative justifications? Conventional BCA relies on standard economic assumptions about human behavior. The most important of these is that people act to maximize their own well-being (subject to the constraints they face). Yet behavioral-economic research provides evidence that people often behave in ways that do not maximize well-being as it is represented by utility functions like those in conventional economic models. Choices seem to be jointly determined by a combination of “real interests” and other factors such as analytic errors, myopic impulses, inattention, passivity, and misinformation. Do these behavioral deviations from the predictions of standard economic models reflect decision-making errors or are standard models oversimplified, ignoring important and legitimate concerns? Surely, both answers are correct: some deviations are due to error, and models are, by design, oversimplified. A more useful question is which of the deviations between behavior and economic models reflect errors that individuals would wish to correct, were they aware of them, and which reflect inadequacies of standard models in describing normatively appropriate behavior? Differences between behavior and standard models drive a wedge between positive and normative justifications for BCA. If people always behave in accordance with standard economic theory, then any policy that satisfies the Kaldor-Hicks compensation test will expand the “social pie” and the central question about BCA would be how to balance efficiency (as measured by aggregate net benefits) against distribution of well-being within society and other concerns. If not, then policies combined with compensation payments that are predicted to yield Pareto improvements may not deliver; affected individuals may not perceive themselves to be better off. How should policy makers and analysts respond when confronted with public preferences that depart from the normative preferences embodied in economic models? Paul Portney posed this question in his 1992 parable, “Trouble in Happyville”: Imagine you are the Director of Environmental Protection for the town of Happyville. There is a naturally occurring contaminant in the town’s drinking water that all of the residents believe is carcinogenic and may account for the towns’ above-average cancer rate. Each resident is willing to pay $1,000 to cover the cost of treatment that will eliminate the contaminant. You have consulted with the world’s leading risk analysts and each has reported that, while one can never be sure, each would stake her professional reputation on the conclusion that this contaminant is benign. You have repeatedly and skillfully communicated these judgments to the citizenry, but each of them still prefers to spend the money to treat the water. What should you do? If you call for the water to be treated, you are knowingly denying each resident the other benefits he could achieve with $1,000 but each resident will believe himself to be better off. If you reject the treatment option, you are knowingly imposing a policy that each resident believes is contrary to his well-being. Implications for Benefit-Cost Analysis If BCA is conceived as a positive exercise, with the goal of determining whether policy consequences satisfy the condition that those who benefit could theoretically compensate those who are harmed, then the objective is to measure benefits and harms exactly as they are perceived by the affected population. When these perceptions conflict with normative models, the normative models are irrelevant. Under this interpretation, analysts should measure individual preferences as accurately as possible. Predicting policy consequences and measuring individual perceptions are scientific questions that are, in principle, susceptible to empirical testing. In addition, this approach respects individual autonomy (consumer sovereignty). If BCA is conceived as a positive exercise, its significance for policy is uncertain. While a practice of choosing policies that satisfy the Kaldor-Hicks compensation test allows for the possibility that everyone in the population will gain, there is no guarantee that such an objective will be achieved and the possibility that other social objectives, such as fair distribution of outcomes or equality of opportunity may be compromised. If BCA is conceived as a normative exercise, then the normative basis must be specified. As the choice of normative basis is a political rather than a scientific question, it seems appropriate for the choice to be made by the relevant political decision makers, though the prospects that they will provide a sufficiently precise statement for analysts to follow seem limited. When using a normative basis, the analyst must determine which parameter values in the BCA are consistent with the corresponding normative model. This may require adjusting empirical estimates to correct for behavioral biases; it is not clear how such adjustments are made. The normative approach assumes the analyst is a better judge of individuals’ well-being than the individuals themselves, and opens her to charges of elitism or paternalism. Many of the questions involved in conducting BCA under a normative justification are not scientific but philosophical and not susceptible to empirical testing, which places in the analyst in more of an advocacy than a scientific role. Nevertheless, benefit-cost analysis that rests on an accepted normative basis is by definition more useful for policy guidance than one that simply predicts if the policy passes the Kaldor-Hicks compensation test. The choice of justification is part of a larger question about the role of representative government: should the government provide the citizenry what the citizenry believes it wants at the moment, much as a direct democracy (or a politician who slavishly follows public-opinion polls) might do, or should it provide leadership, directing the citizenry in a direction it does not yet know (and might never agree) is in its real interest? The tension is nicely encapsulated by the views of two eminent statesmen. Thomas Jefferson (1820): “I know of no safe repository of the ultimate powers of society but the people themselves; and if we think them not enlightened enough to exercise their control with a wholesome discretion, the remedy is not to take it from them, but to inform their discretion by education.” Edmund Burke (1774): “Your representative owes you, not only his industry, but his judgment; and he betrays, instead of serving you, if he sacrifices it to your opinion.” Note Professor Hammitt wrote this article in June 2012, however, due to mistakes on the side of The TSEconomist it was not published in the last issue. We wish to apologize to Professor Hammitt for this inconvenience.
1 Commentaire
Yinghua He is Assistant Professor at Toulouse School of Economics. He holds a PhD from Columbia University. He is specialized in applied microeconomics, labor economics, economics of education and industrial organization. Why is assigning students to public schools and organ donation economics? So ask many people. Although it is cliché, the simple answer lies in the definition of economics: “Economics is a science which studies human behavior as a relationship between ends and scarce means which have alternative uses.” -- Lionel Robbins No doubt that places in public schools and organ donations are scarce, but what is “non-economics” here is that there is no price or monetary transfer involved – public schools are free, and laws usually forbid paying organ donors. This is exactly what makes Al Roth’s studies in market design unique. The goal of market design is to achieve efficient outcomes: Students should be assigned to the schools that they prefer; patients who need an organ transplant should receive a donated organ in time. In a traditional market, this goal can be achieved with the help of a price system. If a student likes a school more, she is willing to pay a higher price, so is a patient in a greater need of an organ transplant. In real life, however, for various reasons, there are many occasions when price or monetary transfer is either forbidden or very limited. The case with the lack of or limited use of monetary transfer is described by the economic term, non-transferable utility. Other examples include allocating doctors to hospitals, housing/dormitory allocation on campus, office allocation, and course allocation. Matching and Stability In the above examples, there are two sides of agents – courses and students, schools and students, or organ donors and patients – whom we want to match together. Note that neither side can be divided into pieces: A seat in a course or a school cannot be shared by two students; neither can a kidney be shared by multiple patients. Hence agents on each side have indivisibility, and yet they are heterogeneous. Otherwise, if all agents are the same, there is little room to have better allocations. One may have noticed that we have been mixing two different scenarios. In office/housing/course allocation, offices, houses and courses do not make active decisions nor do they rank agents on the other side. This is so called one-sided matching which includes the everyday product market. The product itself does not care about who its owner is. There is yet another very different setting, two-sided matching, where both sides make active decisions and rank the opposite side. Hospitals care which doctors they hire, while doctors have preferences for hospitals; schools rank students; and both men and women actively select whom they marry. In addition to non-transferable utility, this area of research assumes that agents are rational such that they know their best interests and behave accordingly. After participating in a matching procedure, or a matching mechanism, agents end up with some allocations where agents from two sides matched with someone from the opposite side. The key concept in matching is stability. As we assume rational agents, a “good” allocation or matching outcome should be such that no agent has incentives to leave her current match. More formally, the allocation is stable if there is no single agent who can obtain any gains from leaving her current match either to stay unmatched or to form a new match with some agent from the opposite side who will be strictly better off in the new match. One-Sided Matching One-sided matching is common in real life. The markets for consumer products, such as TV, computers, and cars, are examples. These products are usually allocated to people based on a price system, or the competitive market mechanism. Whoever can pay the price of the product gets the product. According to the first fundamental theorem of welfare economics, the allocation of a competitive market is Pareto efficient. Namely, by switching to a new allocation, no one can be made better off without someone being made worse off. Therefore, the allocation is also stable. This may explain why the market mechanism prevails in real life. One may wonder why we cannot always use the market mechanism in one-sided matching. In many cases, it is exactly because we are not allowed to use a price system or monetary payments. For example, selling human organs or selling children is forbidden in almost all countries. This is what Roth (2007) describes as repugnancy, and many of his contributions are designing the market given repugnancy as a constraint. A key solution in one-sided matching with non-transferable utility is the top-trading cycle algorithm (TTC). Consider a set of agents and a set of indivisible objects, say houses, without using side-payments. Each agent initially owns one house and each house cannot be shared. The basic idea is that we may repeatedly find a subset of agents who could all obtain their preferred houses by swapping among themselves. More formally, the TTC works as follows: Step 1: Each agent “points to” the owner of her favorite house. Since there are a finite number of agents, there is at least one cycle. Each agent in a cycle is assigned the house of the agent she points to and removed from the market with her assignment. If there is at least one remaining agent, proceed with the next step. Step t: Each remaining agent points to the owner of her favorite house among the remaining ones. Every agent in a cycle is assigned the object of the agent she points to and removed from the market with her assignment. If there is at least one remaining agent, proceed with the next step. Shapley and Scarf (1974) proved that the TTC algorithm, being attributed to David Gale, always produces a stable allocation. Abdulkadiroğlu and Sönmez (1999) later generalized the model to allow for the possibility that some agents do not initially own any objects, while some objects have no initial owner. This model of one-sided matching may seem a bit too abstract, but Roth, Sönmez and Ünver (2004) notice that it can be extended to kidney exchange which is an important real-world situation. Suppose that there are two imaginary figures in our world: François Hollande, who is in need of a kidney transplant, and his girlfriend, Valérie Trierweiler, who is willing to donate one of her kidneys to him. Unfortunately, as it usually happens in real world, the two’s blood types do not match, and therefore Trierweiler’s kidney cannot be used in Hollande’s body. In a competitive market, Hollande could just purchase a kidney if he could afford it, or if not, Trierweiler could sell one of her kidneys and use the money to buy a compatible one for Hollande. For various reasons, this kind of transaction is repugnant and cannot happen. This is then very similar to the house allocation problem, except that Hollande “owns” a house in which he is not able to live. More importantly, there are many patients in a situation like Hollande – it may not be difficult to find someone to give you a kidney, but you have to be very lucky to have a compatible donor. The idea of TTC can then be used to improve the lives of patients in this case. The information of all patient-donor pairs as Hollande and Trierweiler can be gathered, and figuratively each pair points to a pair who is willing to donate a compatible kidney. Whenever we can find a cycle among these pairs, we may perform the transplant for the patients in the cycle. A simple example is illustrated below. Suppose that there is another incompatible patient-donor pair Nicolas Sarkozy - Carla Bruni. Each pair alone is in a desperate situation, but, luckily, the two pairs together can form a cycle and lives can be saved. One important limitation of these cycles is that it is a small probability to find pairs among whom we can carry out the swapping. Bruni is willing to give Hollande her kidney only if Sarkozy is receiving a compatible kidney from Trierweiler – This happens rarely. Fortunately, there are also extremely generous people who are willing to donate their kidneys to anyone in need. In this case, as there are altruistic donors, we can form kidney exchange chains which make more transplants possible. For example, the following figure shows 30 transplants initiated by an altruistic donor.[1] Note that the last patient, Donald C. Terry Jr., bottom right, does not need to find a donor for the first donor, Rick Ruzzamenti, upper left, or for anyone Al Roth and his co-authors have been and still are working on issues related to human organ transplants. One of the most important tasks is to make patients, donors, and also hospitals to participate in this kind of transplant chains. The thicker the market is, i.e., the more people participate, the more chains can be formed. Certainly, the idea of the TTC algorithm can be applied to other cases of one-sided matching with non-transferable utility. Two-sided matching Two-sided matching is different in that both sides of agents make active decisions. College/university admission, school choice, and worker-firm match are common examples. Usually, worker-firm matching is considered in a transferable utility setting where firms face no restrictions when paying workers. It is therefore studied in another literature, search and matching theory, pioneered by Peter A. Diamond, Dale T. Mortensen, and Christopher A. Pissarides, the three Nobel laureates in 2010. In contrast, the tuition fees that public universities and public schools can charge students are highly regulated if not free. Therefore, a non-transferable utility setting is more appropriate. The most important mechanism in two-sided matching is the Gale-Shapley’s deferred acceptance mechanism (DA). It was introduced by Gale and Shapley in a model of marriage market where every man has a strict preference ranking over all women, and vice versa. The DA can be set up in two alternative ways: either men propose to women, or women propose to men. In the latter case, the process begins with each woman proposing to the man she likes the best. Each man then looks at the different proposals he has received if any, retains what he regards as the most attractive proposal but defers from accepting it and rejects the others. The women who were rejected in the first round then propose to their second-best choices, while the men again keep their best offer and reject the rest. This continues until no women want to make any further proposals. As each of the men then accepts the proposal he holds, the process comes to an end. First thing one has to remember is that the success of this mechanism in the marriage market has not been supported by scientific evidence. This may explain that there is no marriage market organized in this way. Nonetheless, the DA mechanism has been proven very useful in many other settings. For example, when matching doctors with hospitals in the US, the National Resident Matching Program (NRMP) uses an algorithm which Roth (1984) found to be essentially equivalent to the employer-proposing DA mechanism. More importantly, recall that the key concept in matching is stability. Gale and Shapley theoretically prove that the DA mechanism always produces stable allocations. In a series of papers, Roth documents the evolution of the market for new doctors in the U.S. and in different regions in the U.K. and argues convincingly that a stable algorithm is the key of a functioning market. The following table from Roth (2002) highlights the importance of stability. There are only two cases where an unstable mechanism is still in use. These studies help us understand those markets better and have opened the door to Roth’s participation in actual market design since the 1990s. Besides, the usefulness of the theory on market design has been further proven in the setting of school choice. In most urban areas in the US, students have the option to choose among the available public schools. Again, there is no price system and students and schools are heterogeneous. In Boston, New York City, and many other places, with the help from Roth and his co-authors, the mechanism used to allocate students to public schools has been reformed toward the DA or the TTC.
Market Design in Daily Life Market design in the absence of a price system, or with non-transferable utility, is still a growing field. Al Roth is still working on the design of markets for human organs, or more broadly the design of markets with repugnance. Besides, as society progresses, new challenges arise as well. For example, due to the increase in female labor force participation, more and more doctors jointly apply for positions as couples at hospitals. We need some new design in the DA mechanism to accommodate these new applications. This need for market design is everywhere in daily life. In France, the AFFELNET system (Affectation des élèves par internet) is used in Paris and other cities to assign students from collége to lycées. It uses a version of school-proposing DA mechanism where students are allowed to rank 6 schools. In 2010 alone, the system allocated 16,500 students to 109 lycées in Paris. Another example is allocating M1 students to M2 programs at TSE. In all these cases, monetary payments are not used. The knowledge from Shapley, Roth, and other contributors in this field can help us better understand how these markets work and improve their functioning. It is important to note that results from this field of research are by no means a substitute to the traditional market mechanism where the price system allocates resources. Instead, they are rather complementary, given repugnance is still prevalent in society. While we keep our fingers crossed for a competitive market, we can improve the market for public schools and organ donation even without a price system if the economist works as engineer. [1] The details are available in the New York Times report, “60 Lives, 30 Kidneys, All Linked,” which also has the copyright of this figure. The article is available at: http://www.nytimes.com/2012/02/19/health/lives-forever-linked-through-kidney-transplant-chain-124.html References and Further Readings: Al Roth’s blog: http://marketdesigner.blogspot.com Al Roth’s Webpage: http://www.stanford.edu/~alroth/ or http://kuznets.fas.harvard.edu/~aroth/alroth.html Abdulkadiroglu, Atila and Tayfun Sonmez. 1999. "House Allocation with Existing Tenants." Journal of Economic Theory, 88(2), 233-60. Economic Sciences Prize Committee of the Royal Swedish Academy of Sciences, 2012, “Stable Allocations and the Practice of Market Design,” Scientific Background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2012. Gale, David E. and Lloyd S. Shapley. 1962. "College Admissions and the Stability of Marriage." American Mathematical Monthly, 69(1), 9-15. Roth, Alvin E. 1984. "The Evolution of the Labor Market for Medical Interns and Residents: A Case Study in Game Theory." The Journal of Political Economy, 92(6), 991-1016. Roth, Alvin E. 2002. "The Economist as Engineer: Game Theory, Experimentation, and Computation as Tools for Design Economics." Econometrica, 70(4), 1341-78. Roth, Alvin E. 2007. "Repugnance as a Constraint on Markets." Journal of Economic Perspectives, 21(3), 37-58. Roth, Alvin E.; Tayfun Sönmez and M. Utku Ünver. 2004. "Kidney Exchange." Quarterly Journal of Economics, 119(2), 457-88. Shapley, Lloyd and Herbert Scarf. 1974. "On Cores and Indivisibility." Journal of Mathematical Economics, 1(1), 23-37. Marc Lebourges is currently Head of European and Economic Regulation, France Telecom Corporate Regulatory Affairs. He was previously marketing director of France Telecom domestic wholesale division. He started as a researcher in operation research and network management studies at the Centre National d’Etudes des Télécommunications and moved to France Telecom’s strategic planning division were he was involved in the preparation and the implementation of interconnection, access and universal service regulation in France and in the elaboration of France Telecom’s Internet strategy. Marc Lebourges is a graduate from the Ecole Nationale Supérieure des Télécommunications and holds a PhD in Computer Science from Université Paris VI. I noticed you hold a PhD in Computer Science. How did your career unfold to lead you to becoming an Economist?
I specialised in Operations Research so I worked in, what was at the time, the Research Department in Operations Research, Queuing Theory and Network Modelling. I have a degree in Telecommunication Engineering and Computer Science in the domain of the areas just mentioned, so I became an expert in Network and Traffic Modelling. I then switched to Strategic and Regulatory Studies and Economics because the topic of Economic Network Modelling became very important for Network Cost calculation in the context of the opening of telecommunications to competition. I came to the Strategic and Economic Research Division of France Telecom during the debate of opening up the network to competition. In particular I was working on financing universal access. I came to be an expert in Network and Electronic Communication Economics by having contact with academics, being part of regulatory debates and then being involved in competition law issues. After some years in more operational positions in marketing within France Telecom, I had the opportunity to come back to Regulatory Affairs and organise a team of economists, some dedicated to practical issues, others to research questions. Are there many professional Economists working at France Telecom? There are some: on one hand, there are economics graduated who are working at France Telecom no longer as Economists and on the other, people like me who have a different background but who tend to work in economic studies. To be more specific, I work in the domain of Regulatory and Competition issues with people of different backgrounds: for example, there are engineers, who work on network cost calculations or service cost calculations, as well as management graduates from business schools and PhDs in economics. Often they need to have legal knowledge as well as economic expertise, as we work very closely with legal experts. There are also people working in the Marketing and Strategic Division who may have an economic background. There are people in R&D developing new types of business models such as long-term business models and the evolution of the service. France Telecom also has Economists in the Accounting and Finance Division. If France Telecom wants to raise debt for instance, we have Macroeconomists checking, very precisely, the market conditions around the world in order to decide where best to issue the debt. Is there a graduate scheme for young economists to grow through the company, or is it the case, as you’ve said, that people are either taken from other areas within the company or taken as experienced hires? In general, Economists tend to study firms from the outside rather than from the inside. It is more management science within the company, so being an Economist is not a standard position within France Telecom. Standard positions are in Marketing, Engineering, Finance and Accounting. Economists are needed due to regulatory debates and competition law cases. Economic analysis is used to inform debate on public policy issues. It is also important that we can justify our position to our competitors and various forms of Public Authorities. During the Business Talk you spoke briefly about the long relationship between France Telecom and TSE. Could you give us some insight into this partnership? France Telecom has supported the IDEI since its creation in 1991. At this time the issue of regulation of networks was very important and this was the expertise of Jeans-Jaques Laffont and Jean Tirole. In particular Access Pricing and Interconnection Issues have been a very important field of research in our sector. We were involved in trying to encourage the use of high quality economic research and analysis to create a rational regulation and market structure. We were looking for economic research to enlighten political decisions in this field. Currently, Bruno Jullien and Wilfried Sand-Zandman are researching a Net Neutrality issue, which will have significant practical consequences. It regards how the network should allocate its cost; either to Internet users or content providers. The problem is that there is traffic sent to subscribers that is not required by these users. However, even for the traffic that is required by users, the quantity of traffic generated is not controlled and cannot be predicted by the user or the provider. Therefore if the end user is to be charged for this traffic we have an inefficiency in the market because they do not have full information regarding controlling their consumption. In certain circumstances the content provider doesn’t have full information either so there is a theoretical issue of how to price in a two-sided market when neither side of the market has the correct information to be efficiently priced. On the one side it is the cost of information, on the other, it is the demand of the value information, both of which are not available. It appears difficult, but we need a theoretical approach to decide on a satisfactory pricing system. The working paper by Jullien and Wilfried Sand-Zandman has been presented at different conferences and helps to analyse this problem. Are you working with other researchers at the IDEI? Our relationship with the IDEI is made up of three main parts. Firstly, researchers at the IDEI keep us informed on economic developments on critical issues that concern France Telecom. Secondly, we ask the researchers at the IDEI to verify that our reasoning in our own economic analysis, used in our on going debate with public authorities, is sound or not. Finally, there is a list of research topics that they have been working on. For example whether local authorities should subsidise or develop their own networks in parallel or in complement to privately owned networks. The IDEI have also generalised the issue to a more generic contract theoretical problem. Other topics include Access Pricing and Margins Squeeze. To what extent should margins squeeze be banned or not and what are the economic consequences of this? Is brand loyalty strategy welfare enhancing or not? Should companies be allowed to develop them or not? Wilfried and Bruno have also done work in these areas. Is the IDEI your main academic partner and to what extent do you use Economic Consultancies? The IDEI is our main academic partner. It is useful to society to use rigorous economic analysis to inform public decisions. Therefore it is in our interest to contribute to the fact that this economic analysis is shared within the academic community. We also have partnerships with more sector specific economic specialists, for instance with the laboratories of Telecom Paristech and of École Polytechnique. We also use consultancies but it is usually on a short-term basis. For example, we use their economic expertise on a case when we are in court with a competitor or competition authority. With consultancies we are defending our case while with academic partners we are trying to help inform public decisions. What proportion of your team’s time is spent conducting research? Where do you source the majority of your data? I am working on operational issues and also have management duties so I do not personally do research. In my team there are 4 Operational Economists, 2 Research Economist’s and 3 PhD students. So that gives you the proportion of the activity. In terms of data, we buy outside data from public databases (such as Informa) and we also use our own anonymised data from the company (from our subscribers etc.). I would say its 50:50. On behalf of the TSE students, thank you for your time, it has been very insightful! |
Archives
Octobre 2016
Categories
Tout
|